Models
AI Models & APIs in India
200+ frontier and open-source models on one API, priced in Indian rupees. Use Claude, GPT, Gemini, Llama, DeepSeek, Qwen and more, and pay in your local currency, on one bill.
Every model, by provider
Anthropic Claude
Claude API| Model | Input / 1M | Output / 1M | Context | Best for | Processed in | |
|---|---|---|---|---|---|---|
Claude Fable 5.1 Anthropic | ₹1,009.40 | ₹5,047.00 | 1M | Long-running agents and the hardest problems | Global | |
| Cached input | ₹25.24₹1,261.75 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹504.70 | ₹2,523.50 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about Claude Fable 5.1 | ||||||
Claude Opus 5 Anthropic | ₹504.70 | ₹2,523.50 | 1M | Products already built on Opus 5 | Global | |
| Cached input | ₹50.47₹630.88 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹252.35 | ₹1,261.75 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about Claude Opus 5 | ||||||
Claude Opus 5.5 Anthropic | ₹403.76 | ₹2,018.80 | 1M | Complex coding and knowledge work | Global | |
| Cached input | ₹20.19₹504.70 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹201.88 | ₹1,009.40 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about Claude Opus 5.5 | ||||||
Claude Sonnet 5.5 Anthropic | ₹201.88 | ₹1,009.40 | 1M | Everyday coding and documents | Global | |
| Cached input | ₹20.19₹252.35 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹100.94 | ₹504.70 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about Claude Sonnet 5.5 | ||||||
Claude Haiku 4.5 Anthropic | ₹100.94 | ₹504.70 | 200K | Live chat and support agents | Global | |
| Cached input | ₹10.09₹126.18 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹50.47 | ₹252.35 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about Claude Haiku 4.5 | ||||||
OpenAI GPT
OpenAI API| Model | Input / 1M | Output / 1M | Context | Best for | Processed in | |
|---|---|---|---|---|---|---|
GPT-6 Astra OpenAI | ₹1,009.40 | ₹5,047.00 | 1.05M | Hard maths, science and agent tasks | Global | |
| Cached input | ₹100.94₹1,261.75 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹504.70 | ₹2,523.50 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about GPT-6 Astra | ||||||
GPT-5.6 Sol OpenAI | ₹403.76 | ₹2,018.80 | 1.05M | Long coding tasks | Global | |
| Cached input | ₹40.38₹504.70 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹201.88 | ₹1,009.40 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about GPT-5.6 Sol | ||||||
GPT-6.1 Sol OpenAI | ₹201.88 | ₹1,009.40 | 1.05M | Production coding agents | Global | |
| Cached input | ₹10.09₹252.35 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹100.94 | ₹504.70 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about GPT-6.1 Sol | ||||||
GPT-6 Sol OpenAI | ₹201.88 | ₹1,009.40 | 1.05M | Products already built on GPT-6 Sol | Global | |
| Cached input | ₹20.19₹252.35 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹100.94 | ₹504.70 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about GPT-6 Sol | ||||||
GPT-5.3-Codex OpenAI | ₹176.65 | ₹1,413.16 | 400K | Terminal coding agents | Global | |
| Cached input | ₹17.66 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about GPT-5.3-Codex | ||||||
GPT-5.6 Luna OpenAI | ₹20.19 | ₹121.13 | 1.05M | High-volume extraction | Global | |
| Cached input | ₹2.02₹25.24 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹10.09 | ₹60.56 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about GPT-5.6 Luna | ||||||
GPT-5.4 nano OpenAI | ₹20.19 | ₹126.18 | 400K | Classification and tagging at scale | Global | |
| Cached input | ₹2.02 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹10.09 | ₹63.09 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about GPT-5.4 nano | ||||||
gpt-oss-120b OpenAI | ₹19.11 | ₹76.44 | 131K | Maths, science and logic | Global | |
| Cached input | ₹9.56 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about gpt-oss-120b | ||||||
GPT-6 Luna OpenAI | ₹10.09 | ₹50.47 | 1.05M | High-volume extraction and routing | Global | |
| Cached input | ₹1.01₹12.62 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹5.05 | ₹25.24 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about GPT-6 Luna | ||||||
gpt-oss-20b OpenAI | ₹8.92 | ₹31.85 | 131K | Quick reasoning steps | Global | |
| Cached input | ₹3.82 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about gpt-oss-20b | ||||||
TTS-1 OpenAI | ₹1,514.10 characters | — | — | Real-time voice replies | Global | |
| Standard rates only. Everything about TTS-1 | ||||||
Whisper large-v3-turbo OpenAI | ₹1,415.56 audio seconds | — | — | Transcripts and captions | Global | |
| Standard rates only. Everything about Whisper large-v3-turbo | ||||||
text-embedding-3-large OpenAI | ₹13.12 | — | 8K | Multilingual search and retrieval | Global | |
| Batch | ₹6.56 | Free | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about text-embedding-3-large | ||||||
text-embedding-3-small OpenAI | ₹2.02 | — | 8K | Low-cost search and retrieval | Global | |
| Batch | ₹1.01 | Free | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about text-embedding-3-small | ||||||
Google Gemini
Gemini API| Model | Input / 1M | Output / 1M | Context | Best for | Processed in | |
|---|---|---|---|---|---|---|
Gemini 4 Argon Google | ₹201.88 | ₹1,009.40 | 1.05M | Finance, legal and research work | Global | |
| Cached input | ₹10.09 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹100.94 | ₹504.70 | Results within 24 hours, for work nobody is waiting on. | |||
| Priority | ₹363.38 | ₹1,816.92 | Faster service, for something a person is sitting in front of. | |||
| Everything about Gemini 4 Argon | ||||||
Gemini 3.1 Pro Google | ₹201.88 | ₹1,211.28 | 1.05M | Video, audio and long PDFs | Global | |
| Cached input | ₹20.19 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹100.94 | ₹605.64 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about Gemini 3.1 Pro | ||||||
Gemini 3.8 Flash Google | ₹75.71 | ₹378.53 | 1.05M | Coding agents at a mid-range price | Global | |
| Cached input | ₹7.57 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Batch | ₹37.85 | ₹189.26 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about Gemini 3.8 Flash | ||||||
Gemma 4 31B Google | ₹17.84 | ₹50.96 | 256K | Documents, forms and images | Global | |
| Batch | ₹8.92 | ₹25.48 | Results within 24 hours, for work nobody is waiting on. | |||
| Everything about Gemma 4 31B | ||||||
Meta Llama
| Model | Input / 1M | Output / 1M | Context | Best for | Processed in | |
|---|---|---|---|---|---|---|
Llama 3.3 70B Meta | ₹37.33 | ₹287.03 | 128K | Drafting and document Q&A | Global | |
| Standard rates only. Everything about Llama 3.3 70B | ||||||
Llama 3.1 8B Meta | ₹7.00 | ₹11.20 | 128K | Very low-cost tagging and routing | Global | |
| Standard rates only. Everything about Llama 3.1 8B | ||||||
DeepSeek
| Model | Input / 1M | Output / 1M | Context | Best for | Processed in | |
|---|---|---|---|---|---|---|
DeepSeek-V4-Pro DeepSeek | ₹152.88 | ₹305.76 | 1M | Coding across large repositories | Singapore | |
| Cached input | ₹22.93 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about DeepSeek-V4-Pro | ||||||
DeepSeek-V4.1-Flash DeepSeek | ₹19.11 | ₹76.44 | 1.05M | Low-cost coding and tool use | Singapore | |
| Cached input | ₹1.91 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about DeepSeek-V4.1-Flash | ||||||
Alibaba Qwen
| Model | Input / 1M | Output / 1M | Context | Best for | Processed in | |
|---|---|---|---|---|---|---|
Qwen 3.8 Max Alibaba | ₹216.58 | ₹649.74 | 1M | Frontier-level coding agents | Singapore | |
| Cached input | ₹28.67₹318.50 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about Qwen 3.8 Max | ||||||
Qwen 3.8 27B Alibaba | ₹44.59 | ₹267.54 | 262K | Low-cost coding agents | Singapore | |
| Cached input | ₹11.47₹79.63 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about Qwen 3.8 27B | ||||||
Qwen 3.8 27B (Colombo) Alibaba | ₹29.40 | ₹147.00 | 131K | Data that must stay in Sri Lanka | Colombo, Sri Lanka | |
| Standard rates only. Everything about Qwen 3.8 27B (Colombo) | ||||||
Qwen 3.8 Flash Alibaba | ₹16.24 | ₹50.90 | 1M | High-volume coding help | Singapore | |
| Cached input | ₹1.83₹25.48 to write | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about Qwen 3.8 Flash | ||||||
More models
| Model | Input / 1M | Output / 1M | Context | Best for | Processed in | |
|---|---|---|---|---|---|---|
Grok 4.6 xAI | ₹403.76 | ₹1,211.28 | 500K | Science and technical questions | Global | |
| Cached input | ₹50.47 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about Grok 4.6 | ||||||
Kimi K3 Moonshot AI | ₹363.09 | ₹1,815.45 | 1.05M | Web research agents | Singapore | |
| Cached input | ₹34.40 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about Kimi K3 | ||||||
GLM-5.3 Zhipu AI | ₹124.85 | ₹392.39 | 1M | Coding agents and security work | Singapore | |
| Cached input | ₹32.10 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about GLM-5.3 | ||||||
MiMo-V2.5-Pro Xiaomi | ₹55.42 | ₹110.84 | 1M | Long autonomous coding sessions | Global | |
| Cached input | ₹0.4586 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about MiMo-V2.5-Pro | ||||||
MiniMax-M3 MiniMax | ₹38.22 | ₹152.88 | 1.05M | Coding with screenshots and video | Global | |
| Cached input | ₹7.64 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about MiniMax-M3 | ||||||
GLM-5.3-Flash Zhipu AI | ₹19.11 | ₹63.70 | 1M | Cheap, high-volume coding agents | Global | |
| Cached input | ₹3.82 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about GLM-5.3-Flash | ||||||
MiMo-V2.5 Xiaomi | ₹17.84 | ₹35.67 | 1M | Audio, video and image analysis | Global | |
| Cached input | ₹0.3567 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about MiMo-V2.5 | ||||||
Hunyuan Hy3 Tencent | ₹13.38 | ₹55.42 | 256K | Coding and office assistants | Global | |
| Cached input | ₹4.20 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about Hunyuan Hy3 | ||||||
Solar Pro 4 Upstage | ₹11.47 | ₹45.86 | 524K | Long documents, without guessing | Global | |
| Cached input | ₹2.29 | — | Repeat a long prompt prefix and those tokens bill at this rate. Automatic. | |||
| Everything about Solar Pro 4 | ||||||
Jev TypeSafe AI | ₹4.24 | Free | 64K | Grading and guardrail checks | Global | |
| Standard rates only. Everything about Jev | ||||||
Prices are our published US dollar rates converted at ₹98.00 to the dollar, our rate for October — the same rate your card is charged.
One API. One bill in Indian rupees.
Instead of a separate account and card with every AI provider, teams in India get one API key and one bill, charged in INR at our published rate for the month. Usage can be tracked across models, projects and team members from one account. See pricing in INR.
Where your requests go
Requests from India go through our gateway in Bangalore. The table above says where each model is processed: those marked Singapore run there, one runs in Colombo, and the rest on international infrastructure — none in India today.
Head to head
The pairs people actually weigh against each other, with prices side by side and a straight answer on which to use.
Claude Fable 5.1 vs GPT-6 AstraClaude Opus 5.5 vs GPT-6.1 SolGemini 3.1 Pro vs Claude Opus 5.5GPT-6.1 Sol vs GPT-6 AstraClaude Sonnet 5.5 vs GPT-6.1 SolClaude Sonnet 5.5 vs Claude Opus 5.5Claude Opus 5.5 vs Claude Opus 5Claude Sonnet 5.5 vs Gemini 3.8 FlashGrok 4.6 vs GPT-6.1 SolGPT-5.3-Codex vs Claude Sonnet 5.5GPT-6 Luna vs Claude Haiku 4.5GPT-6 Luna vs Gemini 3.8 FlashDeepSeek-V4.1-Flash vs GPT-6 LunaGLM-5.3-Flash vs Qwen 3.8 FlashDeepSeek-V4-Pro vs Kimi K3DeepSeek-V4-Pro vs GLM-5.3Qwen 3.8 Max vs DeepSeek-V4-ProMiMo-V2.5-Pro vs Kimi K3Kimi K3 vs Claude Sonnet 5.5gpt-oss-120b vs Llama 3.3 70BQwen 3.8 27B vs Gemma 4 31BQwen 3.8 27B (Colombo) vs Qwen 3.8 27BMuse Spark 1.3 vs Muse Spark 1.3 Contributortext-embedding-3-large vs text-embedding-3-small
Questions
Can I access Claude, GPT and Gemini APIs from India?
Yes. Roar AI gives teams in India access to models from Anthropic, OpenAI, Google and other providers through one API and one key.
Can I pay for AI API usage in Indian rupees?
Yes. Card top-ups are charged in INR at our published rate for the month, and the prices on this page are converted at that same rate.
Are AI models hosted in India?
Not today. The models marked Singapore are processed there, one model runs in Colombo, Sri Lanka, and the rest run on international infrastructure. Each model page says where.
Do I need a different API integration for each model?
No. Every model is reached through the same Roar AI gateway and the same SDK call, so switching models is a one-word change.
Every model,
priced in INR
One API key for every model, on one bill in your local currency.